Our proposed research will examine the determinants of individual-level savings rates, asset allocation, and asset accumulation, with a focus on 401 (k) plans. We will study the impact on savings outcomes of 401 (k) plan design, economic incentives (such as employer matching contributions), how information is presented, and intertemporal preferences. Our research will also test many of the hypotheses from the behavioral economics literature. Several of the patterns in our data suggest that psychological factors play an important role in determining investment and savings choices. These include procrastination, status quo bias, framing, salience effects, peer effects, heuristic decision-making, mental accounting, and more general bounded rationality effects. Because our data are well suited to evaluate the importance of many of the psychological factors noted above, our research is likely to have broad implications for the development of the field of behavioral economics. Many of our proposed projects will exploit administrative data from a large 401 (k) plan administrator. Much existing research on individual saving relies on self-reported information in datasets such as the CPS, SIPP, MRS, and the SCF. Our data are valuable because the information on 401 (k) choices and outcomes is much more detailed and less prone to measurement error than household survey data. Administrative data also provides us with information on the context in which those choices were made (e.g., the constraints the plan design imposed). Our database currently includes up to 8 annual cross- sectional observations from 1998 to 2005 on almost 6 million individuals at 137 large firms. We also have detailed data on every contribution rate and asset allocation transaction at 8 of these firms-170 million transactions during the past 9 years. Over the next few years, we will increase the number of companies in the database and the time period over which these companies are observed.